A computer-implemented recommendation may recommend items of interest to a user browsing an electronic network resource, such as a retail web site or a library web site. The recommendations generated by such a computer-implemented service may be customized to a particular user based on information known about the user. One common application for a recommendation service involves recommending goods and services (“items”) for potential purchase to users shopping on a retail website. For example, the recommendation service may recommend items (e.g., books, compact discs, videos, apparel, accessories, personal electronic devices, etc.) to a user based on a profile that has been developed for the user. For instance, the profile may include information regarding the user's purchase history, rental history, product viewing history, item tagging activities, wish lists and/or other historical or behavioral aspects. Recommendation services are also commonly used for recommending web sites, articles, music and video files, television shows, restaurants and other types of items to users.
While advertising and recommending products and services via web sites and other network resources is a relatively recent technological advance, commercial television has long been used to advertise items. However, targeting strategies for advertising on commercial television have traditionally not been based on personalized or individual recommendations.